Re: PyGtk and gtk-3.0 compatibility
- From: Robert Park <rbpark exolucere ca>
- To: John Stowers <john stowers lists gmail com>
- Cc: python-hackers-list <python-hackers-list gnome org>, gtk-devel-list gnome org, pygtk <pygtk daa com au>, Murray Cumming <murrayc murrayc com>
- Subject: Re: PyGtk and gtk-3.0 compatibility
- Date: Mon, 8 Nov 2010 09:15:53 -0700
On Sun, Nov 7, 2010 at 2:24 PM, John Stowers
<john stowers lists gmail com> wrote:
> I don't know what I plan to do. Thoughts appreciated.
This seems like a fairly cut-and-dry situation to me. PyGTK is
officially considered "legacy support" only, all efforts should be
placed on making PyGObject as good as it can possibly be.
The idea of a backwards-compatible PyGTK-3 was noble, but if it's not
possible to maintain compatibility, then don't waste your time.
Just consider the most likely use-case for a backwards-incompatible PyGTK3:
Alice is an application developer who has written a non-trivial PyGTK2
application. She is informed that PyGTK2 is no longer supported and
that she'll need to port her app to stay current. She looks at the
available options and sees that she can either port to PyGTK3 or
PyGObject. She doesn't really know what PyGObject is, but based on the
name alone determines that PyGTK3 is the successor to PyGTK2. She
begins porting her app to PyGTK3 and eventually completes her port
successfully. Now what? She just spent a bunch of time porting to an
evolutionary dead-end and is left with something that is still old,
broken, and unsupported. Now she has to spend a bunch MORE effort
porting to PyGObject.
I think the community as a whole would be FAR better served by having
not just better PyGObject code, but better documentation for PyGObject
(including porting HOWTOs).
In my humble opinion, that is ;-)
--
http://exolucere.ca
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]